QuassLabs API MCP server
com.quasslabs/apiquasslabs.comstreamable-httpv0.4.0Scanned
30+ deterministic dev and delivery tools: linting, metrics, diagrams, model advice, AI-text checks.
37 tools: 1 can delete or cancel things, 2 do something that can't be undone, 30 change things and 4 only read.
- Reads private data · 1
- Reads text others write · 6
- Can send data out · 2
- 11 findings, 1 high
Tools listed without sign-in. Remote posture
With Anlyon
1 tool here can delete or cancel things and 2 do something that can't be undone. Govern them.
- Anlyon lists the tools and adopts none.
- Each tool you adopt gets approvals and receipts.
- Your agent never holds the key.
A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.
What the QuassLabs API MCP server can do
The QuassLabs API MCP server is published by quasslabs.com, a domain the MCP Registry verified. It has 37 tools. It can delete or cancel things (1 tool: rate_call), do things that cannot be undone, such as sending, paying or publishing (data_transform and send_feedback) and create or change data (30 tools: alignment_gate, diagram_contrast, humanizer_score, saas_metrics and 26 more). 4 tools only read.
1 tool reads private data (log_analyzer); 6 tools bring in text other people can write (alignment_gate, blm_land_lookup, dependency_check and 3 more); 2 tools can send data out of your system (data_transform and send_feedback). On its own it has all three parts of a prompt-injection exfiltration path: private data, text others write, and a way out.
The fixed checks matched 11 findings, 1 of them high severity (annotation-conflict, credentials-mention and sensitive-path). Each one comes with the text it matched.
Its remote endpoint lists its tools without sign-in, and its tools changed 2 times since Anlyon Radar first saw them.
Last scanned Nov 9, 2026, version 0.4.0.
Tools
Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.
37 tools
Deletes or cancels 1Claude Code: deny
rate_callRate a specific prior metered call - thumbs up or down, plus an optional comment - keyed by the request_id that call's response reported.No findings
Why it is in this group
- The name "rate_call" says "call": a word for a change.
- The server marks it destructive (destructiveHint: true).
Description, as the server wrote it Rate a specific prior metered call - thumbs up or down, plus an optional comment - keyed by the request_id that call's response reported. Works for any metered tool, so you can tell us when a result was wrong, unhelpful, or missed something. Resubmitting the same request_id overwrites your previous rating rather than adding a second one. Auth required; not billed - rating a call you already paid for should not cost more.
Annotations the server declares
- title: Rate a prior call
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: true
- openWorldHint: false
Can't be undone 2Claude Code: ask
data_transformFive deterministic data utilities behind one tool, chosen by operation.Can send data out1 finding
Why it is in this group
- The input addresses people (input.to): what it sends cannot be taken back.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- outboundThe input takes a destination (input.to): what the tool writes goes there.
Findings
- Warn
The server marks "data_transform" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The input addresses people (input.to): what it sends cannot be taken back.
Description, as the server wrote it Five deterministic data utilities behind one tool, chosen by operation. csv_normalize: sniffs the delimiter, strips the BOM and normalises line endings, snake_cases headers (duplicates renamed), pads or trims ragged rows, drops empty rows and optionally duplicate rows, and returns CSV or JSON rows (max 200,000 cells, 1,000 columns, 64 KB per cell). json_shape: infers a JSON Schema (draft 2020-12) from 1-1,000 samples, with required, enum and format only where every sample supports them (max depth 32, 200,000 nodes). json_extract: evaluates up to 50 paths from a safe JSONPath subset ($, .key, ['key'], [n], [-n], [*], .*, ..key); filters, scripts, slices and unions are rejected, and each path returns at most 10,000 matches. unit_convert: exact rational conversion within one dimension (length, mass, volume, temperature, time, data, speed, area, pressure, energy), rounded to 12 significant digits, single or up to 1,000 per batch. text_dedupe: exact, normalized (NFKC, casefold, whitespace and edge punctuation) or near-duplicate (3-word-shingle Jaccard) removal over up to 50,000 strings, 2 MB total (near mode 5,000). 2 MB request body. Caller content is never executed or evaluated and nothing is fetched. Deterministic, no LLM.
Annotations the server declares
- title: Deterministic data utilities
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
send_feedbackSubmit free-text feedback about the API - a bug, an idea, a docs gap, a billing question, or anything else.Can send data outNo findings
Why it is in this group
- The name "send_feedback" says "send": a word for sending, and what is sent cannot be taken back.
What it can reach
- outboundThe name "send_feedback" says "send": it sends or publishes something to other people.
- outboundThe input takes a URL (input.endpoint): a request to a URL can carry data to whoever runs it.
Description, as the server wrote it Submit free-text feedback about the API - a bug, an idea, a docs gap, a billing question, or anything else. Missing a tool? Use kind tool_request to tell us what you needed, with an optional tool_description. Tied to your key so we can follow up, but not a support-ticket SLA. Auth required; not billed.
Annotations the server declares
- title: Submit product feedback
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: false
Changes things 30Claude Code: ask
acceptance_criteria_validatorQA advisor: validate acceptance criteria quality.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it QA advisor: validate acceptance criteria quality. POST a list of AC (or a story) and get per-AC problems (vague/untestable, non-atomic, no observable outcome) + a missing-coverage flag if there is no failure-path AC, plus a 0-100 score and pass/needs_work verdict. Tells you exactly what to fix. Deterministic, no LLM.
Annotations the server declares
- title: Acceptance criteria validator
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
accountEverything about your QuassLabs API account, chosen by `action`.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Everything about your QuassLabs API account, chosen by `action`. Free: no action costs credits. "signup" (no API key needed): request a free key for an email address. A confirmation link is emailed; a human opens it on quasslabs.com to see the key, which is never returned to the agent. The key comes with 300 free credits a month. Always answers status check_email, whether or not the address already has a key. Needs email, accept_terms (true, after the human has agreed to https://quasslabs.com/api/terms) and terms_version. "balance" (API key required): what the calling key can spend - purchased credit_balance, free_credits left this UTC month, when they reset, plan and key_id. Daily free-tier caps still apply, so a free call can return 429 free_tier_busy while free_credits is above 0. Use it before a batch of paid calls or after a 402 insufficient_credits. "buy_credits" (API key required): a Stripe Checkout link for a human to pay in a browser; starter pack $5 for 500 credits, pro pack $20 for 2500. Credits land on this key once payment succeeds. "checkout_status" (no API key needed): whether a checkout session from buy_credits has been paid and its credits granted.
Annotations the server declares
- title: Account: sign up, balance, buy credits
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
actions_auditStatic security audit of GitHub Actions workflow files.1 finding
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Findings
- Info
The description mentions "credentials".
What this check looks forMatched text
description: "...s: inherit, actions/checkout persisting credentials, self-hosted runners on public reposito..."
Description, as the server wrote it Static security audit of GitHub Actions workflow files. Flags template injection (attacker-controlled github.* contexts such as a PR title or branch name expanded inside run: or actions/github-script), pull_request_target and workflow_run triggers and the pwn-request pattern of checking out the pull request head under them, actions and reusable workflows not pinned to a full 40-character commit SHA, container images not pinned by digest, missing or write-all GITHUB_TOKEN permissions, secrets used under untrusted triggers or passed with secrets: inherit, actions/checkout persisting credentials, self-hosted runners on public repositories, curl or wget piped to a shell, hardcoded credentials (reported by kind and line, never echoed), jobs with no timeout-minutes, and caching in release or tag-publish workflows. Up to 50 files of 256 KB each, 1 MB request body; YAML anchors and aliases are rejected. Nothing referenced is fetched, so pinned SHAs are not checked to exist and called actions are not audited. Rules follow zizmor and GitHub's Actions hardening guide. Deterministic, no LLM.
Annotations the server declares
- title: GitHub Actions workflow security audit
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
agent_card_lintStatic lint of an A2A (Agent2Agent) Agent Card, the JSON an agent publishes at /.well-known/agent-card.json.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Static lint of an A2A (Agent2Agent) Agent Card, the JSON an agent publishes at /.well-known/agent-card.json. Checks conformance with A2A v1.0.1 and the 0.3 field names (both dialects accepted): required fields present and typed, URLs absolute https (http allowed for localhost), input and output modes that look like MIME types, skills with id, name, description and tags, duplicate skill ids, and security requirements that reference declared schemes. Scores discoverability: skill descriptions too short (under 20 chars) or too long (over 1,024) to route on, near-duplicate skills, missing examples or provider. Flags prompt-injection phrasing aimed at the model reading the card and hidden zero-width, bidi or tag characters (reported as U+XXXX, never echoed raw). Optionally checks an x402 payment declaration: every accepts entry needs scheme, network, payTo, asset, resource (https) and a sane atomic amount; x402 findings are reported separately and do not affect the card score. 512 KB request; first 200 skills, 50 items per list, 100 x402 entries. URLs are parsed, never fetched, and card signatures are not verified. Deterministic, no LLM.
Annotations the server declares
- title: A2A Agent Card linter
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
alignment_gateRun an AI system prompt (and optional agent config) through a 45-criterion alignment/safety rubric.Reads text others write1 finding
Why it is in this group
- The description's first verb is "run": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Findings
- Warn
The server marks "alignment_gate" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":false,"openWorldHint":true}; The description's first verb is "run": a word for a change.
Description, as the server wrote it Run an AI system prompt (and optional agent config) through a 45-criterion alignment/safety rubric. Returns a letter grade (A-F), numeric score, per-flag remediation, and a pass/fail verdict.
Annotations the server declares
- title: Alignment/safety rubric scorer
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
commit_hygieneDeterministic commit-message hygiene checker: conventional-commit format, vague-subject detection, subject-length limit, a same-message secret-pattern scan (AWS keys, private key h…2 findings
Why it is in this group
- The name "commit_hygiene" says "commit": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Findings
- Info
The description mentions "credentials".
What this check looks forMatched text
description: "...tokens, inline password=/secret= style credentials - checked against the message text only..."
- Warn
The server marks "commit_hygiene" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "commit_hygiene" says "commit": a word for a change.
Description, as the server wrote it Deterministic commit-message hygiene checker: conventional-commit format, vague-subject detection, subject-length limit, a same-message secret-pattern scan (AWS keys, private key headers, Slack/GitHub tokens, inline password=/secret= style credentials - checked against the message text only, never a diff), and an oversized-commit warning from caller-supplied insertions/deletions. POST one commit or a batch (max 200). Deterministic, no LLM, no network.
Annotations the server declares
- title: Commit message checker
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
convert_diagramMake Mermaid diagrams portable.1 finding
Why it is in this group
- The name "convert_diagram" says "convert": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Findings
- Warn
The server marks "convert_diagram" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "convert_diagram" says "convert": a word for a change.
Description, as the server wrote it Make Mermaid diagrams portable. Takes HTML containing <div class="mermaid"> blocks (or a raw Mermaid source) and returns the same document with every diagram pre-rendered to inline SVG and the Mermaid CDN runtime stripped. A CDN-backed diagram only renders if the reader can fetch and run the runtime; inline SVG needs neither, so the output survives offline readers, email clients, Notion/Confluence (which strip <script>), and PDF export. Rendering happens on our infrastructure; diagram content is not sent to any third party.
Annotations the server declares
- title: Mermaid to inline SVG
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
coverage_checkScope-completeness advisor.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Scope-completeness advisor. POST a brief (or requirements[]) + your stories[] and get a coverage score + the gaps (requirements no story covers), orphan stories (possible scope creep), and missing structural buckets (auth/CRUD/error-states/tests). Tells you what the scope is missing. Deterministic, no LLM.
Annotations the server declares
- title: Requirements coverage check
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
create_uploadGet a presigned S3 PUT URL to upload a file (policy doc, config, log, up to 5MB) that a file-ingesting endpoint then reads by upload_id.No findings
Why it is in this group
- The name "create_upload" says "create": a word for a change.
Description, as the server wrote it Get a presigned S3 PUT URL to upload a file (policy doc, config, log, up to 5MB) that a file-ingesting endpoint then reads by upload_id. Two-step: call this, PUT your file to the returned url, then pass upload_id to the consuming endpoint. Uploads auto-expire in 1 day and are readable only by your key. Auth required; not billed.
Annotations the server declares
- title: Get a presigned upload URL
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: false
defect_triageBug-triage advisor.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Bug-triage advisor. POST one bug report (or a list) and get severity (critical/high/medium/low), category, whether it has repro steps, and what info is missing (no repro / no expected-vs-actual / no environment). For a list: priority ranking + likely-duplicate clusters. Deterministic, no LLM.
Annotations the server declares
- title: Bug report triage
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
dependency_checkCheck dependency manifests.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Check dependency manifests. Price: free by default; with live: true, 2 credits up to 50 packages, +1 per further 50, max 5 (200 packages). Default: offline hygiene lint for unpinned, wildcard or unbounded versions, missing lockfiles, git/URL deps, non-https indexes, install scripts, dependency-confusion names, plus a static snapshot of famous compromised versions. live: true adds, per package, known vulnerabilities from OSV.dev (exact versions only), latest version, licence and deprecation from deps.dev, and one verdict: PASS, REVIEW or FAIL (FAIL at fail_on severity or above, a denied licence, or a compromised version). Use this for manifests you can paste; use supply_chain_scan for a whole repo ZIP or MCP configs. Send exactly one of manifests (1-20 files) or upload_id + filename. Reads package.json, package-lock.json, requirements.txt, pyproject.toml, Pipfile, go.mod, Gemfile, composer.json, pubspec.yaml. Nothing is installed or run. Bearer key required. Returns hygiene findings, a 0-100 score and grade A-F, and with live: true a live section with the verdict, per-package results and source attribution. Upstream outages are refunded (503 upstream_unavailable).
Annotations the server declares
- title: Dependency check: hygiene, plus live vulnerabilities
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
diagram_contrastDiagram quality QA.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Diagram quality QA. Catch and fix the things that ruin diagrams: (1) unreadable label contrast - POST a mermaid/SVG or HTML diagram, get every WCAG-AA failure (incl. mermaid <style>-block label colors, with unresolvable fills flagged honestly, not false-passed) + auto-corrected colors; (2) harden a mermaid SOURCE before it renders - rewrites every fill/label pair and remaps fills no label color can rescue, which scanning cannot do because a per-node style line overrides the theme block; (3) HTML-looks-fine-but-prints-broken - checks for the print-color-adjust/page-break gotchas and returns the fix CSS; (4) arrows crossing objects - returns the ELK-renderer config + layout rules that minimize crossings. Pure deterministic - no AI, no rate limit.
Annotations the server declares
- title: Diagram contrast and print QA
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
estimate_sanityDeterministic classifier for story/task hour estimates against benchmarks: flags each estimate in_range, borderline, over, under, or unbenchmarked.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Deterministic classifier for story/task hour estimates against benchmarks: flags each estimate in_range, borderline, over, under, or unbenchmarked. Supply either an explicit {min, max} range per type (benchmarks) or a list of historical actual hours per type (historical) that this endpoint derives a mean/stdev variance band from - explicit ranges win when both are given for the same type. No benchmark numbers are built in; a type with neither input is reported unbenchmarked, never guessed at. POST one estimate or a batch (max 200). Deterministic, no LLM.
Annotations the server declares
- title: Estimate sanity check
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
humanize_textRewrite text to remove AI-writing tells.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Rewrite text to remove AI-writing tells. POST prose and get it back with the mechanical tells fixed (AI vocabulary swapped for plain words, filler phrases stripped, em-dash overuse reduced, significance inflation deflated) plus a change log, before/after ai-ness scores, and a residual list of tells that need human judgment (not auto-rewritten). Code fences, inline code, URLs, and quotes are never touched. Deterministic, no LLM. Sister to humanizer_score (the detector).
Annotations the server declares
- title: AI-tell rewriter
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
humanizer_scoreDetect AI-writing tells in text.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Detect AI-writing tells in text. POST prose and get a 0-1 ai_ness score + band (low/medium/high) + the EXACT patterns that fired (AI vocabulary, significance inflation, promotional language, filler phrases, negative parallelism, rule-of-three, em-dash overuse) with sample spans. Explainable, not an opaque classifier - every point is a named, located pattern. Deterministic, no LLM. The rewrite is a separate step.
Annotations the server declares
- title: AI-writing tell scorer
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
log_audit_creatorGenerate a concrete audit-logging spec for an app.1 finding
Why it is in this group
- The name "log_audit_creator" says "log": a word for reading.
- The description's first verb is "generate": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Findings
- Warn
The server marks "log_audit_creator" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "log_audit_creator" says "log": a word for reading.
Description, as the server wrote it Generate a concrete audit-logging spec for an app. POST your feature list (and optional compliance targets: soc2/hipaa/gdpr/pci) and get a full audit-event catalog (namespaced event names, actor/target/required fields, PII flags, retention), a canonical event envelope schema, a storage plan (append-only, access control, alerting), and per-framework compliance notes. A dev team can implement it directly. Deterministic, no LLM. Retention/compliance figures are common-practice defaults to verify with counsel.
Annotations the server declares
- title: Audit-log event catalog generator
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
mcp_tool_lintStatic review of MCP tool definitions, as returned by a server's tools/list, for how well a model can choose and call them.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Static review of MCP tool definitions, as returned by a server's tools/list, for how well a model can choose and call them. Flags missing, too-short or vague descriptions; parameters with no description or type; free-form objects and arrays without items; required fields that are not declared; duplicate, case-colliding or near-identical tools; missing or contradictory annotations (readOnlyHint, destructiveHint and friends); tool-poisoning phrasing that tells the model to override its instructions or hide actions from the user; and hidden zero-width, bidi or tag characters. Also estimates the token cost of the whole tool list. Nothing is fetched and no tool is called; the score is a heuristic for usability, not a security guarantee. Deterministic, no LLM.
Annotations the server declares
- title: MCP tool definition linter
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
model_advisorLLM selection + cost advisor.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it LLM selection + cost advisor. POST a task description (+ optional volume/tokens/constraints like local-only or cheap) and get a recommended model + fallback chain + do-not-use (with reasons), the complexity tier, and per-run + monthly cost. A maintained model+price catalog. Deterministic, no LLM. Tells the AI/user which model to use and what it costs.
Annotations the server declares
- title: LLM model and cost advisor
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
pr_description_lintDeterministic PR description quality linter: flags a missing/too-short body, absent test evidence, absent stated rationale, an unstructured long body, and - only for PRs that look …No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Deterministic PR description quality linter: flags a missing/too-short body, absent test evidence, absent stated rationale, an unstructured long body, and - only for PRs that look like bugfixes (title prefix fix/bug/hotfix/patch, or an explicit is_bugfix flag) - missing repro steps. Optionally requires a linked issue/ticket reference. Single PR only, no batch. Deterministic, no LLM.
Annotations the server declares
- title: PR description linter
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
readability_gradeDeterministic Flesch Reading Ease and Flesch-Kincaid Grade Level for a block of text, plus word/sentence/syllable counts and a plain-language reading-ease band (very_easy through v…No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Deterministic Flesch Reading Ease and Flesch-Kincaid Grade Level for a block of text, plus word/sentence/syllable counts and a plain-language reading-ease band (very_easy through very_confusing). Syllable counting is the standard vowel-group heuristic (same approximation used by textstat and most open readability tools), not a dictionary lookup. Optionally pass target_grade to get an on_target flag (within +/-2.0 grade levels). Different metric family from humanizer_score - this is pure readability statistics, not AI-writing-tell detection. Deterministic, no LLM.
Annotations the server declares
- title: Readability grade
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
regex_lintStatic safety analysis of a regular expression plus an English explanation of what it matches.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Static safety analysis of a regular expression plus an English explanation of what it matches. Parses the pattern into an AST and flags the structures that cause catastrophic backtracking (ReDoS): nested unbounded quantifiers like (a+)+, repeated alternations whose branches can start with the same character like (a|ab)*, adjacent unbounded quantifiers over overlapping character sets like \d+\d+, and implausibly large bounded repeats. The pattern is COMPILED but never EXECUTED against input - running a caller-supplied regex is itself the denial-of-service vector. Detection is a structural heuristic, so a risk of "none" means no known dangerous shape was found, not a proof of safety. Deterministic, no LLM.
Annotations the server declares
- title: Regex ReDoS linter
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
saas_metricsSaaS health advisor.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it SaaS health advisor. POST your MRR/customers (+ optional churn/CAC/margin/growth) and get ARR/ARPA/LTV/CAC/payback/Rule-of-40/NRR/GRR computed, each with a health tag, plus an overall grade + a ranked "what to fix first" focus list. Deterministic math + advisory, no LLM.
Annotations the server declares
- title: SaaS metrics calculator
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
skill_lintStatic lint of an agent instruction file: an Agent Skills SKILL.md (YAML frontmatter plus markdown body) or an AGENTS.md / CLAUDE.md style file.2 findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Findings
- High
The description refers to "~/.ssh", "~/.aws" and ".env".
What this check looks forMatched text
description: "..., bypassing permission prompts, sending ~/.ssh, ~/.aws or .env to a URL, and prompt-in..."
- Info
The description mentions "credentials".
What this check looks forMatched text
description: "...folder deep, hidden or bidi characters, credentials (reported by pattern name and line, nev..."
Description, as the server wrote it Static lint of an agent instruction file: an Agent Skills SKILL.md (YAML frontmatter plus markdown body) or an AGENTS.md / CLAUDE.md style file. For skills, checks the frontmatter against the Agent Skills specification: name present, lowercase-hyphenated, at most 64 characters and free of reserved words; description present, at most 1,024 characters, with no XML tags, written in the third person and saying when to use the skill; license, compatibility, metadata and allowed-tools typed correctly; unknown keys. For both kinds, flags an empty or over-long body (over 500 lines or about 5,000 tokens), skipped heading levels, Windows backslash paths, referenced files missing from an optional sibling file list, references nested more than one folder deep, hidden or bidi characters, credentials (reported by pattern name and line, never echoed), and unsafe instructions such as curl piped to a shell, rm -rf on root or home, git --no-verify, --dangerously-* flags, bypassing permission prompts, sending ~/.ssh, ~/.aws or .env to a URL, and prompt-injection phrases; a match inside a negated or quoted sentence is downgraded to a warning. Text up to 300 KB, 512 KB request. The text is never executed, links are never fetched and referenced files are never read. Deterministic, no LLM.
Annotations the server declares
- title: Agent skill / AGENTS.md linter
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
sprint_healthSprint/delivery health advisor.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Sprint/delivery health advisor. POST 1-24 sprints (committed/completed points, optional per-ticket estimate-vs-actual data) and get computed metrics each with a good/warn/bad health tag: velocity avg + trend + stability, commitment reliability, spillover rate, mid-sprint scope change, estimate accuracy (median + chronic-overrun pct), and WIP concentration per assignee. Plus an overall grade and a ranked what-to-fix-first focus list. Pure math + advisory, no LLM.
Annotations the server declares
- title: Sprint health report
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
sql_migration_lintStatic lint of a SQL migration file (Postgres or MySQL) for operations that lock or rewrite live tables or break running application code.1 finding
Why it is in this group
- The input takes a statement or command to run (input.sql), which can change anything it can reach.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Findings
- Warn
The server marks "sql_migration_lint" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The input takes a statement or command to run (input.sql), which can change anything it can reach.
Description, as the server wrote it Static lint of a SQL migration file (Postgres or MySQL) for operations that lock or rewrite live tables or break running application code. Flags CREATE INDEX without CONCURRENTLY and concurrent index builds inside a transaction, foreign key, CHECK and UNIQUE constraints added without NOT VALID or USING INDEX, SET NOT NULL without a validated CHECK, column type changes, NOT NULL columns added without a default, volatile or serial defaults that rewrite the table, dropped or renamed columns and tables, VACUUM FULL and CLUSTER, DELETE or UPDATE without WHERE, TRUNCATE, a missing lock_timeout, MySQL ALTERs without ALGORITHM/LOCK, and type advice (jsonb, text, identity, timestamptz). Tables created earlier in the same file are treated as new and skipped by lock and rewrite rules. The SQL is tokenized, never executed: no database is contacted, so table sizes and existing column types are unknown, and SQL inside strings, comments and function bodies is not linted. Up to 400 KB of SQL and 5,000 statements. Rules adapted from squawk and strong_migrations. Deterministic, no LLM.
Annotations the server declares
- title: SQL migration zero-downtime linter
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
story_lintDeterministic structural quality linter for user stories - a sibling to ticket_lint but scoped to story-writing concepts any team recognizes: narrative completeness (As-a/I-want-to…No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Deterministic structural quality linter for user stories - a sibling to ticket_lint but scoped to story-writing concepts any team recognizes: narrative completeness (As-a/I-want-to/So-that/Summary), acceptance criteria presence, duplicate/unknown dependency IDs, a "Manage X" story with no CRUD split, list stories missing a pagination mention. An OPTIONAL rate_config unlocks team-specific checks (screen-vs-narrative match, hour-rate-per-screen tolerance, required risk level by story type, milestone-code format, required companion story types per milestone) - omit it and only the universal checks run. No house rates or milestone conventions are hardcoded. POST one story or a batch (max 50). Deterministic, no LLM.
Annotations the server declares
- title: User story linter
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
supply_chain_scanScan a whole repo archive (a ZIP of manifests, lockfiles and MCP server configs) for packages, MCP servers and extensions listed in a known-compromise catalog (Bumblebee threat int…Reads text others write1 finding
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Findings
- Info
The input schema at properties.archive_url.description mentions "credentials".
What this check looks forMatched text
inputSchema.properties.archive_url.description: "...the same kind of ZIP, readable with no credentials (typically presigned), for archives too..."
Description, as the server wrote it Scan a whole repo archive (a ZIP of manifests, lockfiles and MCP server configs) for packages, MCP servers and extensions listed in a known-compromise catalog (Bumblebee threat intel). Use this for a repo, many lockfiles or MCP configs; use dependency_check instead to lint a few pasted manifests for pinning and hygiene. Send exactly one of archive_base64 or archive_url; neither or both is 400 exactly_one_of_archive_base64_or_archive_url. Read-only: the ZIP is sent to QuassLabs (archive_url is fetched server-side), unpacked to a temp dir deleted after the call, never executed or installed. 5 credits. Failed calls are not billed: input, fetch and archive errors (invalid_ecosystems, not_a_zip, unsafe_path_in_archive, too_many_entries, archive_expands_too_large) are rejected before any charge, and 5xx is refunded. Bearer key required. Returns verdict CLEAR or FINDINGS, findings with severity, source_file and remediation, and scanned counts; CLEAR is not a guarantee of safety.
Annotations the server declares
- title: Supply-chain compromise scanner
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
ticket_lintFull Jira-ticket quality linter (whole ticket, not just AC - use acceptance_criteria_validator for the deep AC pass).No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Full Jira-ticket quality linter (whole ticket, not just AC - use acceptance_criteria_validator for the deep AC pass). POST one ticket or a batch (max 50) and get per-ticket findings with severity + concrete fixes: missing/weak summary or description, missing AC or estimate, bugs without repro/expected-vs-actual/environment, missing epic link, points-vs-hours mismatch, oversized estimates to split, urgent-but-thin tickets. Returns score, grade, and a ready/needs_work verdict per ticket plus batch summary. Type-aware rules (story/bug/task/epic/subtask), strict profile available. Deterministic, no LLM.
Annotations the server declares
- title: Jira ticket linter
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
tool_advisorWhich QL-API endpoint fits this task.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Which QL-API endpoint fits this task. POST a task description and get the ranked endpoint(s) that best match, each with a confidence label (high/medium/low). Deterministic keyword matching over this platform's own endpoint catalog, no LLM. Pairs with model_advisor (which itself is one of the recommendable endpoints, so a "which model" task surfaces it) - call them independently and chain the results yourself; optionally pass a prior model_advisor result's rationale in as `context` to bias matching.
Annotations the server declares
- title: QuassLabs endpoint advisor
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
worklog_guardrailWorklog/timesheet policy checker - run a batch of worklog entries through configurable guardrails BEFORE posting them to a time-tracking system.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Worklog/timesheet policy checker - run a batch of worklog entries through configurable guardrails BEFORE posting them to a time-tracking system. Checks: rounding increment (default 0.25h, float-safe), per-entry min/max, per-author daily totals, duplicates, over-estimate alerts (never suggests inflating hours or estimates - flags to PM instead), future dates, weekends, immutable-author edit protection, missing description/ticket. Returns pass/warnings/blocked verdict + per-entry violations with fixes + hour totals by author and date. Caller supplies the reference date - fully deterministic, no LLM.
Annotations the server declares
- title: Worklog policy check
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
Only reads 4Show 4 read-only tools
blm_land_lookupLook up Bureau of Land Management land status for a US coordinate: BLM ownership, jurisdiction, allowed activities (rockhounding, camping), permit requirements, and special designa…Reads text others writeNo findings
Why it is in this group
- The name "blm_land_lookup" says "lookup": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Look up Bureau of Land Management land status for a US coordinate: BLM ownership, jurisdiction, allowed activities (rockhounding, camping), permit requirements, and special designations (ACEC, Wilderness, WSA, NM/NCA).
Annotations the server declares
- title: BLM land status lookup
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
describe_diagramTurn a plain-English description into Mermaid diagram source.Reads text others writeNo findings
Why it is in this group
- The name "describe_diagram" says "describe": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Turn a plain-English description into Mermaid diagram source. Describe the flow, system, schema or timeline you want ("a login flow with an OAuth branch and a failure path") and get back complete, valid Mermaid - no need to know the syntax. The mermaid field of the response is directly usable as the mermaid_source input of convert_diagram, so describe -> render is a two-call pipeline. Generation runs on our infrastructure via AWS Bedrock; your description is not sent to any third-party service beyond that.
Annotations the server declares
- title: Text to Mermaid diagram
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
log_analyzerAnalyze a log file.Reads private dataNo findings
Why it is in this group
- The name "log_analyzer" says "log": a word for reading.
What it can reach
- private-dataThe description's first sentence says "analyze" and "file": it reads files.
Description, as the server wrote it Analyze a log file. POST raw log text (or an upload_id from the create_upload tool, up to 5MB) and get: level breakdown + time range, error clusters (messages normalized so variable ids/numbers collapse into one signature, with counts + first/last seen + a sample), a timeline, detected anomalies (error spikes, retry storms, new error signatures, log gaps/outages, level-ratio shifts) with evidence, and templated recommendations. Handles JSON-lines and plain formats (ISO8601/syslog/CLF/epoch timestamps). Deterministic, single-pass, no LLM.
Annotations the server declares
- title: Log file analyzer
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
openapi_diffCompares a base and a head OpenAPI 3.x spec (JSON object, or a JSON or YAML string) from the API consumer's point of view and lists what would break existing clients: removed opera…No findings
Why it is in this group
- The name "openapi_diff" says "diff": a word for reading.
Description, as the server wrote it Compares a base and a head OpenAPI 3.x spec (JSON object, or a JSON or YAML string) from the API consumer's point of view and lists what would break existing clients: removed operations or 2xx responses, new required parameters or request properties, optional-to-required changes, narrowed request enums and tightened constraints, removed or now-optional response properties, widened response enums, type changes, and newly required security. Additive changes are reported as info. Matches operations by method and path template, so renaming a path parameter is not a change. Local $refs are resolved; remote refs are never fetched. Suggests the semver bump and says whether the head version already reflects it. Deterministic, no LLM.
Annotations the server declares
- title: OpenAPI breaking-change diff
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.
Exfil surface
Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.
12 paths in this server alone from private data, through text someone else wrote, to a way out. Denying 1 private-data tool breaks every one.
- Reads private data1
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write6
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out2
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
{
"permissions": {
"deny": [
"mcp__api__log_analyzer"
]
}
}Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.
Remote posture
What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.
Looked at Nov 20, 2026, 18:23 UTC, unchanged since Nov 13, 2026. No credentials were sent and no tool was called. What is looked at.
- Reachable
- yes
- HTTP status of an anonymous initialize
- 200
- tools/list served without credentials
- yes
- Tools that change things, listed without credentials
- 33
- Protocol version
- 2025-06-18
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- not observed
- Authorization servers
- not observed
- Legacy HTTP+SSE transport
- no
- Dynamic client registration
- not observed
- Client ID metadata documents
- not observed
- HighThe whole server
The endpoint served tools/list without credentials, and the list has 33 tools that change things (1 destructive, 2 irreversible, 30 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://api.quasslabs.com/mcp: tools/list returned 37 tools to an anonymous client, including rate_call (destructive), data_transform (irreversible), send_feedback (irreversible), alignment_gate (write), diagram_contrast (write) and 28 more
Posture changes
- Nov 13, 2026
- Tools that change things, listed without credentials: 26 to 33
Findings
Matches of fixed checks, with the text they matched.
- Highskill_lint
The description refers to "~/.ssh", "~/.aws" and ".env".
What this check looks forMatched text
description: "..., bypassing permission prompts, sending ~/.ssh, ~/.aws or .env to a URL, and prompt-in..."
- Warnalignment_gate
The server marks "alignment_gate" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":false,"openWorldHint":true}; The description's first verb is "run": a word for a change.
The server marks "log_audit_creator" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "log_audit_creator" says "log": a word for reading.
- Warncommit_hygiene
The server marks "commit_hygiene" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "commit_hygiene" says "commit": a word for a change.
- Warndata_transform
The server marks "data_transform" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The input addresses people (input.to): what it sends cannot be taken back.
The server marks "sql_migration_lint" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The input takes a statement or command to run (input.sql), which can change anything it can reach.
Show 5 more findingsShow fewer
- Warnconvert_diagram
The server marks "convert_diagram" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "convert_diagram" says "convert": a word for a change.
- Infocommit_hygiene
The description mentions "credentials".
What this check looks forMatched text
description: "...tokens, inline password=/secret= style credentials - checked against the message text only..."
- Infoactions_audit
The description mentions "credentials".
What this check looks forMatched text
description: "...s: inherit, actions/checkout persisting credentials, self-hosted runners on public reposito..."
- Infoskill_lint
The description mentions "credentials".
What this check looks forMatched text
description: "...folder deep, hidden or bidi characters, credentials (reported by pattern name and line, nev..."
The input schema at properties.archive_url.description mentions "credentials".
What this check looks forMatched text
inputSchema.properties.archive_url.description: "...the same kind of ZIP, readable with no credentials (typically presigned), for archives too..."
History
Changes to its tool definitions and to its remote endpoint, newest first.
PostureRemote posture changed. Tools that change things, listed without credentials: 26 -> 33.View diff
- Tools that change things, listed without credentials: 26 to 33
Tools5 added, 1 changed.+5 added~1 changedView diff
- Added:actions_auditChanges things
- Added:agent_card_lintChanges things
- Added:data_transformCan't be undone
- Added:skill_lintChanges things
- Added:sql_migration_lintChanges things
- Changed:dependency_checkChanges thingsdescriptioninputSchemaannotations
Tools7 added, 4 removed, 25 changed.+7 added−4 removed~25 changedView diff
- Added:accountChanges things
- Added:create_uploadChanges things
- Added:humanize_textChanges things
- Added:mcp_tool_lintChanges things
- Added:openapi_diffOnly reads
- Added:send_feedbackCan't be undone
- Added:supply_chain_scanChanges things
- Removed:buy_credits
- Removed:feedback
- Removed:humanize
- Removed:uploads
- Changed:acceptance_criteria_validatorChanges thingsinputSchemaannotations
- Changed:alignment_gateChanges thingsinputSchemaannotations
- Changed:blm_land_lookupOnly readsannotations
- Changed:commit_hygieneChanges thingsinputSchemaannotations
- Changed:convert_diagramChanges thingsannotations
- Changed:coverage_checkChanges thingsinputSchemaannotations
- Changed:defect_triageChanges thingsinputSchemaannotations
- Changed:dependency_checkChanges thingsdescriptioninputSchemaannotations
- Changed:describe_diagramOnly readsannotations
- Changed:diagram_contrastChanges thingsannotations
- Changed:estimate_sanityChanges thingsinputSchemaannotations
- Changed:humanizer_scoreChanges thingsannotations
- Changed:log_analyzerOnly readsdescriptioninputSchemaannotations
- Changed:log_audit_creatorChanges thingsinputSchemaannotations
- Changed:model_advisorChanges thingsinputSchemaannotations
- Changed:pr_description_lintChanges thingsinputSchemaannotations
- Changed:rate_callDeletes or cancelsdescriptioninputSchemaannotations
- Changed:readability_gradeChanges thingsannotations
- Changed:regex_lintChanges thingsannotations
- Changed:saas_metricsChanges thingsinputSchemaannotations
- Changed:sprint_healthChanges thingsinputSchemaannotations
- Changed:story_lintChanges thingsinputSchemaannotations
- Changed:ticket_lintChanges thingsinputSchemaannotations
- Changed:tool_advisorChanges thingsinputSchemaannotations
- Changed:worklog_guardrailChanges thingsinputSchemaannotations
Install
Pinned to version 0.4.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http api https://api.quasslabs.com/mcpclaude mcp add --transport http api https://api.quasslabs.com/mcpRun in a terminal.
Claude Code permissions
Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.
{
"permissions": {
"allow": [
"mcp__api__blm_land_lookup",
"mcp__api__log_analyzer",
"mcp__api__openapi_diff",
"mcp__api__describe_diagram"
],
"ask": [
"mcp__api__alignment_gate",
"mcp__api__diagram_contrast",
"mcp__api__humanizer_score",
"mcp__api__saas_metrics",
"mcp__api__acceptance_criteria_validator",
"mcp__api__coverage_check",
"mcp__api__defect_triage",
"mcp__api__model_advisor",
"mcp__api__tool_advisor",
"mcp__api__humanize_text",
"mcp__api__dependency_check",
"mcp__api__log_audit_creator",
"mcp__api__ticket_lint",
"mcp__api__story_lint",
"mcp__api__estimate_sanity",
"mcp__api__pr_description_lint",
"mcp__api__commit_hygiene",
"mcp__api__readability_grade",
"mcp__api__regex_lint",
"mcp__api__mcp_tool_lint",
"mcp__api__actions_audit",
"mcp__api__agent_card_lint",
"mcp__api__data_transform",
"mcp__api__skill_lint",
"mcp__api__sql_migration_lint",
"mcp__api__sprint_health",
"mcp__api__worklog_guardrail",
"mcp__api__create_upload",
"mcp__api__send_feedback",
"mcp__api__account",
"mcp__api__supply_chain_scan",
"mcp__api__convert_diagram"
],
"deny": [
"mcp__api__rate_call"
]
}
}Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as api. Review it before you use it.
Capability tags behind 9 rules
- mcp__api__alignment_gate untrusted-input
- mcp__api__blm_land_lookup untrusted-input
- mcp__api__dependency_check untrusted-input
- mcp__api__log_analyzer private-data
- mcp__api__data_transform outbound
- mcp__api__send_feedback outbound
- mcp__api__account untrusted-input
- mcp__api__supply_chain_scan untrusted-input
- mcp__api__describe_diagram untrusted-input
Pin
Keep the decision: fail CI when the tools change, follow every change, or show a badge.
Pin in CI
Fail a build when the tools you reviewed are not the tools the server serves today.
curl -s "https://radar.anlyon.com/api/v1/servers/com.quasslabs/api/verify?hash=35948d3f5c01dd9f7a0fcffb85b9ee6abcdf33f995fb90a342203b729e636eea" | jq -e '.changed == false'Pinned hash 35948d3f5c01, from Nov 9, 2026, 02:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/com.quasslabs/api/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/com.quasslabs/api)Source: MCP Registry · repository. Synced Nov 9, 2026, 02:17 UTC.
QuassLabs API MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the QuassLabs API MCP server safe?
- No scan can promise that. What its definitions show: 37 tools, 33 of which can change, send or delete something, and 11 findings (1 high) from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.
Can the QuassLabs API MCP server delete data?
- Yes. 1 tool can delete, cancel or revoke something: rate_call.
Can the QuassLabs API MCP server do things that cannot be undone?
- Yes. 2 tools can send, pay, publish or merge: data_transform and send_feedback.
What tools does the QuassLabs API MCP server have?
- 37 tools: 1 destructive, 2 irreversible, 30 write, 4 read. rate_call, data_transform, send_feedback, alignment_gate, diagram_contrast, humanizer_score, saas_metrics, acceptance_criteria_validator and 29 more.
Does the QuassLabs API MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the QuassLabs API MCP server send data to outside services?
- 2 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): data_transform and send_feedback.
Has the QuassLabs API MCP server changed its tools?
- Yes, 2 times since Anlyon Radar started tracking it, most recently on Nov 9, 2026. Every change is on this page and in its Atom feed.